Please use this identifier to cite or link to this item: http://bura.brunel.ac.uk/handle/2438/9648
Title: Genetic folding for solving multiclass SVM problems
Authors: Mezher, MA
Abbod, MF
Keywords: Classification;Evolutionary algorithm;Genetic folding;GF;Kernel function;SVM
Issue Date: 2014
Citation: Applied Intelligence, 41(2): 464-472, (17 April 2014)
Abstract: Genetic Folding (GF) algorithm is a new class of evolutionary algorithms specialized for complicated computer problems. GF algorithm uses a linear sequence of numbers of genes structurally organized in integer numbers, separated with dots. The encoded chromosomes in the population are evaluated using a fitness function. The fittest chromosome survives and is subjected to modification by genetic operators. The creation of these encoded chromosomes, with the fitness functions and the genetic operators, allows the algorithm to perform with high efficiency in the genetic folding life cycle. Multi-classification problems have been chosen to illustrate the power and versatility of GF. In classification problems, the kernel function is important to construct binary and multi classifier for support vector machines. Different types of standard kernel functions have been compared with our proposed algorithm. Promising results have been shown in comparison to other published works.
URI: http://link.springer.com/article/10.1007%2Fs10489-014-0533-1
http://bura.brunel.ac.uk/handle/2438/9648
DOI: http://dx.doi.org/10.1007/s10489-014-0533-1
ISSN: 0924-669X
1573-7497
Appears in Collections:Dept of Electronic and Computer Engineering Research Papers

Files in This Item:
File Description SizeFormat 
Fulltext.doc178.5 kBMicrosoft WordView/Open


Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.